The gram-positive bacterium Staphylococcus epidermidis is the most common pathogen in hospital-acquired infections. The costs related to infections caused by Staphylococcus epidermidis in the hospital setting are enormous (more than 1 billion dollars/year) and represent a major health care burden. S. epidermidis, usually a commensal of human skin, may cause septicemia or endocarditis in patients undergoing immunosuppressive therapy, premature newborns, or injection drug users. However, most infections caused by S. epidermidis occur after the insertion of indwelling devices such as catheters or prosthetic heart valves. In these cases, the ability of S. epidermidis to form biofilms represents the most important virulence determinant. In a biofilm, the bacteria are dramatically less susceptible to antibiotic treatment and to attacks by human immune defenses. For these reasons, S. epidermidis infections are very difficult to eradicate. We propose that drugs preventing and/or targeting biofilm formation will be of extraordinary use in anti-staphylococcal therapy, as they will enable the immune system to cope with an infection and increase the efficiency of common antibiotics. To provide the scientific basis for the development of drugs interfering with biofilm formation, we are investigating the molecular biology, biochemistry, and epidemiology of biofilm formation in S. epidermidis. This includes studying specific factors contributing to biofilm formation, their regulation, and the interaction of biofilm-forming S. epidermidis strains with the host. Investigation of biofilm formation of Staphylococcus epidermidis has been divided into 3 projects: 1. Control of biofilm formation by global regulatory systems We have constructed deletion mutants in two of the three major global regulatory systems using a standard biofilm-forming strain. The third deletion mutant is about to be constructed. The deletion mutants shall be analyzed in vitro for biofilm formation on plastic and in vivo by animal models. An animal model of subcutaneous catheter infection in rabbits is being developed. Vectors for the expression analysis of two of the three regulatory systems in biofilms using confocal laser scanning microscopy have been constructed. These vectors carry different fluorescent protein derivatives as reporters. We have begun comparing the agr (accessory gene regulator) locus, which is one of the global regulatory loci and responsible for quorum sensing, in strains from catheter infections and from healthy individuals. These analyses include sequencing of the agr locus, expression of biofilm factors, biofilm formation in vitro, and further phenotypes. Results from these studies will give valuable information about the role of the quorum sensing system agr in infection in S. epidermidis. According to the sequence of the peptide pheromone that is produced by the agr locus, several agr subgroups have been found previously in S. aureus. Very recently, data obtained from DNA sequencing suggested the presence of different subgroups in S. epidermidis, too. We are constructing agr deletion mutants of all S. epidermidis agr subgroups to determine the peptide sequence of the respective pheromones and to compare the function of agr in the different S. epidermidis agr subgroups. 2. Biochemistry of the exopolysaccharide PIA (polysaccharide intercellular adhesin), identification and characterization of enzymes responsible for the production, alteration, and degradation of PIA PIA is the main determinant responsible for intercellular adhesion, which is considered to be the second step in biofilm formation. Studies have shown that presence of the ica genes responsible for PIA expression is correlated with catheter infection in humans. ica mutants show a distinctly decreased ability to cause catheter infection in animal models. PIA is a homopolymer of partially de-acetylated N-acetyl-glucosamine units. We want to investigate the influence of de-acetylation on biofilm formation and on the impact on the effect of the human immune defense. The degree of PIA acetylation determines the charge of PIA. Differences of the charge of PIA, and consequently of bacterial surface charge, are thought to dramatically influence these characteristics. A deletion mutant of the putative de-acetylase gene of S. epidermidis (icaB) is being constructed. We have constructed vectors for the over-expression of IcaB and made antisera against IcaB. This allowed us to localize IcaB and propose a new model for IcaB processing to its mature form. A further goal is to find enzymes that degrade PIA polymers. By analyzing the S. epidermidis genome, we have found a good candidate for this activity. We have cloned the corresponding gene in over-expression vectors and are about to construct a deletion mutant. We have also constructed a transposon bank comprising about 20000 mutants that are currently being screened for biofilm formation. Mutants altered in biofilm formation shall be further characterized. By this approach we believe to find new factors responsible for biofilm formation and regulation , including enzymes that alter PIA strucure, like PIA-degrading enzymes. 3. The role of the detergent-like peptide delta-toxin and similar peptides in S. epidermidis biofilm formation and inflammation Delta-toxin is a peptide with detergent-like properties encoded within the agr locus of several pathogenic staphylococci. Its role in infection has remained unclear for many years. We have previously shown that it prevents attachment to plastic and influences biofilm formation. Recently, it has also been found as a component of the so-called "phenol-soluble modulin" complex (PSM), which has several inflammatory effects and is composed of the delta-toxin and (at least) two further peptides with similar characteristics. The objectives of this project are to determine the impact of each of these peptides on biofilm formation and on inflammation. Separation of the peptides by chromatographic methods is very difficult, as they are very similar. Chemical synthesis has failed for most of them. We have therefore constructed over-expression vectors for each peptide. We have analyzed expression in the heterologous host S. carnosus, which does not produce biofilm factors and does not produce inflammatory peptides. The latter has been determined by our collaborator in this project, Dr. Klebanoff from the University of Washington at Seattle. Over-expression in S. carnosus was successful. We have developed a purification scheme, which allows us to purify single components of the PSM complex in the milligram range from the supernatant of S. carnosus over-expression strains. This new method is also suitable to purify the entire PSM complex without the need for lengthy purification steps including extraction with phenol. The pure peptides will be analyzed for their impact on biofilm formation and for their interference with the human immune systems using several different assays (at RML and at UW). Together with our collaborators we could also identify a fourth component of the PSM. We have developed a method to rapidly determine the production of each single PSM using LC-MS. Using this method, we are currently investigating the production of each PSM in strains of clinical origin and from healthy individuals to address the importance of PSM in human infection also by epidemiological means.